Abstract
In continuation of our earlier paper [3], we study in detail the teaching assistant model of exact learning. In [3] we show that the exact learnability of some algebraic concept classes can be more sharply classified in this model than in Angluin's model. The teaching assistant model can be seen as an enhancement of Angluin's model. The new ingredient is that apart from the learner and teacher there is a third agent called teaching assistant which acts as an intermediary between the learner and teacher.
In this paper we investigate in detail the learnability of various concept classes in the teaching assistant model. After giving the precise definition of the model, we first discuss the possibility of precisely capturing Angluin's model in the teaching assistant model and prove that this is not possible. We then discuss in detail the power of NP ∩ co-NP and UP ∩ co-UP assistants. Using machine characterizations of learnability with NP ∩ co-NP and UP ∩ co-UP assistants, we show that, information theoretically NP ∩ co-NP and UP ∩ co-UP assistants have the same power. We generalize the notion of teaching dimension [9] and define a notion of weak teaching dimension using which we characterize concept classes that are learnable with NP ∩ co-NP assistants. Finally, with some algebraic concept classes as examples we obtain several separations between the power of various teaching assistant classes.
Preview
Unable to display preview. Download preview PDF.
References
D. Angluin. Queries and concept learning. Machine Learning, 2:319–342, 1988.
D. Angluin. Negative results for equivalence queries. Machine Learning, 5:121–150, 1990.
V. Arvind and N. V. Vinodchandran. The complexity of exactly learning algebraic concepts. Seventh Workshop on Algorithmic Learning Theory, LNAI 1160, 100–112, 1996.
J. L. Balcázar, J. Díaz, and J. Gabarró. Structural Complexity — I & II. Springer Verlag, Berlin Hiedelberg, 1988.
R. Book, T. Long and A. Selman. Quantitative relativization of complexity classes. SIAM Journal on Computing, 13:461–487, 1984.
S. Fenner, L. Fortnow, S. Kurtz. Gap-definable counting classes. Proc. 6th Structure in Complexity Theory Conference, 30–42, 1991.
M. Furst, J. E. Hopcroft and E. Luks. Polynomial time algorithms for permutation groups. Proc. 21st IEEE Symposium of Foundations of Computer Science, 36–45, 1980.
R. Gavaldà. On the power of equivalence queries. EUROCOLT, 193–203, 1993.
S. A. Goldman and M. J. Kearns On the complexity of teaching. Journal of Computer and System Sciences 50: 20–31, 1995.
M. Hall. The Theory of Groups. Macmillan, New York, 1959.
L. Hellerstein, K. Pillaipakkamnatt, V. Raghavan and D. Vilkins. How many queries are needed to learn? Proc. of the 27th Symposium on the Theory of Computating, 190–199, 1995.
O. Watanabe. A formal study of learning via queries. 17th International Colloquium on Automata, Languages and Programming, LNCS 443: 139–152, 1990.
O. Watanabe and R. Gavaldà. Structural analysis of polynomial time query learnability. Mathematical Systems Theory, 27:231–256, 1994.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Arvind, V., Vinodchandran, N.V. (1997). Exact learning via teaching assistants (Extended abstract). In: Li, M., Maruoka, A. (eds) Algorithmic Learning Theory. ALT 1997. Lecture Notes in Computer Science, vol 1316. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63577-7_50
Download citation
DOI: https://doi.org/10.1007/3-540-63577-7_50
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63577-2
Online ISBN: 978-3-540-69602-5
eBook Packages: Springer Book Archive